ACADIAN FRENCH AND LINGUISTIC THEORY1 Ruth King York University ABSTRACT This article identifies some profitable areas of research for Acadian linguis­ tics, in particular, research in sociolinguistics and in grammatical theory. The value of Acadian French data for the testing of theories of language change and for the study of syntactic variation within generative grammar is explored. 1. INTRODUCTION Interest in Acadian French has grown tremendously over the past two decades, and with it, there has been a dramatic increase in articles and monographs on linguistic aspects of these varieties. While the first publication on Acadian French, Pascal Poirier's 'La langue acadienne', appeared in 1884, most publications are considerably more recent. The great major­ ity of the 430 entries in Edward Gesner's 1986 Bibliographie annotée de linguistique acadienne are post-1960 and 65 date from the period 1980-85. In the bibliography, the varieties of Acadian French spoken in all four Atlantic provinces and in Louisiana are well represented.2 Descriptive studies of phonology and vocabulary are especially prominent, understandably so since these are the foci of traditional dialectology. There are also a significant number of mor­ phological studies, as might be expected since most comparative work on Acadian compares it with standard French and Acadian varieties differ considerably from standard French in verb morphology. Obviously documentation of present-day Acadian varieties is both necessary and im­ portant, as evidenced by the continuing value to modern researchers of landmark works such as Geneviève Massignon's 1962 Les parlers français de l'Acadie, a (principally) lexical study based on fieldwork conducted in the mid-1940s. However, in this article I will focus not on the significance of such documentation for the historical record, but on how research on Acadian French may also be important in the development of linguistic theory, specifically sociolinguis- tic theory and grammatical theory.3 In the case of sociolinguistic theory, Acadian communities are quite different from the ones usually investigated by variationists: they are bilingual and they typically do not exhibit the sort of social stratification found, say, in New York City or Norwich. By studying how and by whom linguistic change is implemented in such communities we are able to study the extent to which the Labovian model can account for linguistic innova­ tion and for the diffusion of linguistic change. With respect to grammatical theory, the gram­ mars of varieties of Acadian, while in many respects quite similar, differ from those of other 1 Research on Prince Edward Island varieties of Acadian French has been supported by research grants 410-87-0586, 410-89-0338 and 410-90-0615 from the Social Sciences and Humanities Research Council of Canada. 2 Acadian French refers to varieties of French spoken in North America (principally in New Brunswick, Nova Scotia, Newfoundland and Prince Edward Island in Canada and in Louisiana in the United States) which have their origin in the 'centre-ouest' of France. 3 I will be concerned here with Atlantic Canada Acadian French; for more on the sociolinguistics of Louisiana varieties, see Brown (1988); for more on Louisiana French and grammatical theory, see Brown (1986). 36 Ruth King varieties of French in nontrivial ways. Therefore, they may provide important data for the study of parametric variation. 2. SOCIOLINGUISTIC THEORY It is sometimes remarked that Canadian French is one of the world's language varieties most studied by sociolinguists. Most notably, there have been the pioneering Montreal French studies begun two decades ago by Henrietta Cedergen and Gillian Sankoff and the continuing work on the Cedergren-Sankoff corpus and on more recent Montreal corpora by Cedergren, David Sankoff, Pierrette Thibault and their associates, along with the studies of Ontario French un­ dertaken over the past fifteen years by Raymond Mougeon and his associates, and the investi­ gations of the French spoken in Ottawa-Hull by Shana Poplack and her associates, beginning iri the early 1980s. Over the last decade, sociolinguistic studies of Acadian varieties spoken in all four Atlantic Provinces have also been undertaken (cf. Flikeid 1984 for northeastern New Brunswick, King 1983 for Newfoundland varieties, King and Ryan 1988 for Prince Edward Island varieties, Flikeid 1989 for Nova Scotia varieties). As we might expect, the Acadian studies reveal complex, but structured, organization of linguistic variation, or as Labov has termed it, orderly heterogeneity. They reveal as well tension between maintenance of Acadian linguistic features (carriers of Acadian identity) and linguistic change in the direction of com­ munity-external standards. For instance, in her northeastern New Brunswick study, Flikeid (1984) found evidence of style shifting on the part of younger speakers in the direction of less use of certain well-known phonological features of Acadian, such as palatalized variants of /k / and /g /, but widespread use (across age groups) of certain other features, such as the fronting of low back nasal vowels in stressed, open syllables. A principal finding of Labov and his followers has to do with linguistic changes in progress: change begins with working or lower middle class speakers and then spreads to other social groups (cf. Labov 1966, 1980; Labov, Yaeger and Steiner 1972).4 In his classic New York City study, Labov (1966) developed what has become the standard methodology for sociolin­ guistic studies: informants are ranked on a social class index based on a number of socioeconomic factors and are then divided into social class groupings on the basis of their SEC scores. Such stratification studies have been conducted in a large number of urban contexts, from Philadelphia to Sydney, Australia to Panama City to St. John's, with considerable success. Sociolinguistic studies of Acadian varieties have found the interrelated factors of age, level of education, and level of bilingualism to be the more important social factors in the analysis of linguistic variation and change (cf. Flikeid 1984, King 1983). In general younger, more bilingual, better educated (in French) Acadians speak less conservative Acadian French. For example, in one case of change in progress in the direction of the external standard, the spread of the [w] variant of orthographic -oi- in northeastern New Brunswick, Flikeid (1984) found that age is by far the most important social factor, with younger speakers leading the 4 Labov (1966) develops a curvilinear model of social diffusion of linguistic change wherein the working and lower-middle classes are the innovators. However, Kroch (1978) argues for a linear model in which the dominant force is the upper class's resistance to such innovations. For discussion, and an attempt to reconcile the two approaches, see Guy (1988). While he has in the past rejected the possibility of rapid and sweeping linguistic change, Labov (1991) admits to the possibility of such change but only in face of 'catastrophic social events', (p. 245, my emphasis). Journal of the Atlantic Provinces Linguistic Association 13 (1991) Acadian French and Linguistic Theory 37 change. These results are what we might expect since the small communities studied inten­ sively display little social stratification among their francophone inhabitants. This is not to say that socioeconomic factors are never found to be significant. In one small community currently under study, Abram-Village in Prince Edward Island, the notion of the linguistic marketplace, i.e., of 'how speakers' economic activity, taken in its widest sense, requires or is necessarily associated with competence in the legitimized language' (Sankoff and Laberge 1978: 239), has been found to be of some importance. My account of one case of linguistic variation in Abram-Village (cf. King 1991a), i.e., variation in the use of the well-integrated English lexical borrowing back, is in terms of marketplace ranking. In this specific case there is a negative correlation between use of back and higher marketplace ranking. However, Mougeon and Beniak's studies of a number of francophone communities in Ontario lead them to suggest that, in the case of linguistic change in minority languages, change may not 'proceed in the way described by Labov for monolingual communities, that is, via the introduction of an innovation by an individual speaker or by a small group of speakers belonging to a particular social class, and its subsequent propagation to other speakers of the same class and eventual adoption by speakers of other classes' (Mougeon and Beniak 1991a: 13). The five Ontario communities they have studied in detail, i.e., Hawkesbury, Cornwall, North Bay, Pembroke and Welland, all display social class variation. In their 1991 book they report that in only one case among the many linguistic variables they have investigated is there any­ thing but a loose connection with social class. Rather degree of minority-language-use restric­ tion and level of bilingualism are the key factors.5 Acadian communities are another important testing ground for Mougeon and Beniak's hypothesis that linguistic innovation may involve the autonomous behaviour of one or more speakers. The study of Acadian varieties, including those spoken in areas in which there is clear social stratification (e.g., in urban centres such as Moncton6), in areas in which there is so­ cial differentiaton not readily correlated with socioeconomic factors (e.g., in areas such as Baie Sainte-Marie in Nova Scotia) and in small, relatively homogeneous villages (e.g., in L'Anse-à- Canards in Newfoundland or in Chéticamp in Nova Scotia), allows us to investigate further Mougeon and Beniak's hypothesis. Of course, as Flikeid (1988: 196) points out, this is not an easy undertaking, since 'découvrir les dimensions pertinentes de la structure sociale existante exige des recherches sociologiques, une connaissance intime du milieu et aussi de l'innovation dans les techniques sociolinguistiques utilisés', but it is nevertheless an important one.